# Arrow of Time (Directional Classification) on Something-Something V2 (val)

Source: [Arrow of Time (Directional Classification) on Something-Something V2 (val)](<https://www.sota2.com/research/sota/arrow-of-time-directional-classification-on-something-something-v2-val>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-25T07:39:08.238Z

Evaluation of directional classification performance, specifically focusing on the 'Arrow of Time' aspect, on the validation split of the Something-Something V2 dataset, measured by Accuracy.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Arrow of Time (Directional Classification) |
| Dataset (including any stated split) | Something-Something V2 (val) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Accuracy](<https://www.sota2.com/research/sota/arrow-of-time-directional-classification-on-something-something-v2-val.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| V-JEPA 2 | 89.36 | Not recorded | 6a13d78a6f71c5ae7dc9e8b5 |
| TIME | 88.53 | Not recorded | 6a13d78a6f71c5ae7dc9e8b5 |
| RVM | 81.88 | Not recorded | 6a13d78a6f71c5ae7dc9e8b5 |
| VideoMAEv2 | 71.87 | Not recorded | 6a13d78a6f71c5ae7dc9e8b5 |

### V-JEPA 2

Accuracy: 89.36. Result ID: 6a13fc89fbc373592c0adb7e.

[Reporting paper](<https://www.sota2.com/research/paper/6a13d78a6f71c5ae7dc9e8b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79651e209131cea7d6a4.md>)

[Implementation](<https://github.com/facebookresearch/vjepa2>)

Recorded evaluation configuration: {"Modality":"256×256 Video","Size":"ViT-L","Pre-training Dataset":"Real-World","Training Samples":"22M (88×)","Video Hours Eq.":"≈1.73M (12, 357×)","Pre-training explicitly includes SSv2":"true","Evaluation Protocol":"Linear Probing"}. This record does not assert independent reproduction.

### TIME

Accuracy: 88.53. Result ID: 6a13fc89fbc373592c0adb7f.

[Reporting paper](<https://www.sota2.com/research/paper/6a13d78a6f71c5ae7dc9e8b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a13d78a6f71c5ae7dc9e8b5.md>)

Recorded evaluation configuration: {"Modality":"32×32 Tracks","Size":"ViT-B","Pre-training Dataset":"Synthetic","Training Samples":"0.25M (1×)","Video Hours Eq.":"140 (1×)","Pre-training explicitly includes SSv2":"false","Evaluation Protocol":"Linear Probing"}. This record does not assert independent reproduction.

### RVM

Accuracy: 81.88. Result ID: 6a13fc89fbc373592c0adb7d.

[Reporting paper](<https://www.sota2.com/research/paper/6a13d78a6f71c5ae7dc9e8b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6940e7e5456146936284fcf2.md>)

Recorded evaluation configuration: {"Modality":"224×224 Video","Size":"ViT-B","Pre-training Dataset":"Real-World","Training Samples":"5.68M (23×)","Video Hours Eq.":"≈0.25M (1, 785×)","Pre-training explicitly includes SSv2":"true","Evaluation Protocol":"Linear Probing"}. This record does not assert independent reproduction.

### VideoMAEv2

Accuracy: 71.87. Result ID: 6a13fc89fbc373592c0adb7c.

[Reporting paper](<https://www.sota2.com/research/paper/6a13d78a6f71c5ae7dc9e8b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f171e209131cea43db1.md>)

Recorded evaluation configuration: {"Modality":"224×224 Video","Size":"ViT-B","Pre-training Dataset":"Real-World","Training Samples":"5.36M (27×)","Video Hours Eq.":"≈7,500 (54×)","Pre-training explicitly includes SSv2":"true","Evaluation Protocol":"Linear Probing"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.